TWI364713B - Sequential color reproduction method and apparatus - Google Patents

Description

1364713. EMBODIMENT DESCRIPTION: TECHNICAL FIELD OF THE INVENTION The present invention relates generally to imaging systems; more specifically, to color processing in imaging systems. [Prior Art] A variety of imaging technologies are currently available. For example, an imaging system can use an array of independently addressable pixels such as a deflecting reflective micromirror, a mobile reflective film (such as an IMOD), a liquid crystal cell (such as an LCD), a liquid crystal cell (LCOS), or a radiation tube. Image engine (such as electric tube). The imaging system can also include a passive display or an active display. In the resulting image, such imaging systems may produce an ugly wide image. One of the main imaging systems, especially one of the color-imaging systems, is the color-splitting image, which is also known as color splitting. . In eyeball follow-up and momentary motion, the color separation layer appears as a multi-color image. This artifact is most common in scenes that contain high-contrast spatial transitions such as scrolling white fonts on a black background. And 'different imaging systems have different color spaces or different combinations of primary colors to form a color image. These color spaces and combinations of primary colors may not match the color space or primary color combination of the _ imaging source. For example, an image projected using a particular imaging system may use red, green, and blue primary colors; however, this particular imaging system uses a pixel consisting of red, green, blue, and a fourth color, such as yellow, per pixel. Array. Obviously, displaying an image using pixels that are different in color combination with the input image pixels without proper color processing will result in a poor quality image. 127907.doc 1364713 SUMMARY OF THE INVENTION A method of processing an image will be disclosed in an example. The method includes: deriving a second color image component set of a second color group when receiving an image of a first color image component set having a first color group; wherein the second color group includes a first color group An additional color within a color set; and wherein the derivative of one of the additional color image elements comprises a non-linear transition.

In another example, an imaging system will be disclosed. The system includes: an image processor having an input for receiving an image to be reproduced, wherein the image processing g further comprises: receiving a first color image set having a first color set And a derivative device for deriving a second color image component set of the second color group; wherein the second color group includes an additional color that is not in the first color group; wherein one of the additional color image components The derivative includes a non-linear transformation; and an image engine having an array of pixels that reproduce the input image based on a set of image data derived from the second set of color image elements.

In another example, an imaging system that reproduces an input image will be disclosed. The method includes: an image processing for receiving an input end of an image to be reproduced: wherein the image processor further comprises: an input end for receiving an input image color shirt 7L set; a second mode change module for obtaining an intermediate value from the color component set, a nonlinear data filter for outputting a non-linear change-to-conversion module output signal to the first conversion module; The group input end and the non-linear filter output end are used to obtain a subtraction node and a multiplexer group of the color image component set; and - an output end for outputting the converted color image component set; an image reference 127907. Doc 1364713^ It has a pixel array that reproduces an input image based on an image set derived from the converted set of color image elements. In another example, an apparatus for processing an image will be disclosed. The method comprises: receiving an input image color component set of the input end; a first-conversion module for obtaining the intermediate value from the color component set; - connecting with the first conversion module - using The linear function converts the nonlinear value of the value between the towels, a subtraction node set connected to the input end of the first conversion module and the output of the nonlinear filter, for subtracting one from each of the independent input color image elements A value U proportional to the non-linear conversion value is an output signal set of the output-converted color image component set, the color image component set containing the subtracted color image component. [Embodiment] An improved color management method and an imaging system using the same will be disclosed in the following sections. The color management method converts a primary color image set of an input image into a set of other shirt color image components commensurate with an imaging system displaying the input image, wherein the other color image component is concentrated in at least one The components are converted using a nonlinear conversion function. After conversion, the saturation of the primary color image element of the input image (or the energy it carries) is redistributed between the converted sets of primary color image elements. An additional color band (e.g., white band) different from the primary color of the input image will be introduced in a sequential imaging system; the additional color can be determined based on the color associated with the potentially sequential color separation artifacts associated with the displayed image. The sequential color separation artifacts are eliminated or reduced by redistributing the input image primary component saturation (or energy) between the converted primary color component sets including the additional color image components. For example, in a black and white image that is most perceptible to a color-separated image in a human eye, 'when the extra color is white, the color image component of the displayed image will be reduced or have minimal saturation (or energy); The white image element of the image will have an increased saturation (or energy): thus, the color separation artifact is greatly reduced in the displayed black and white image. This also applies to images other than black and white images, in which case the extra color can be other colors than white. Redistributing saturation and energy can be achieved by calculating additional color components of the image reproduced with the non-linear conversion function; and correspondingly reducing the saturation (or energy) of other primary components. The calculation of additional color components is based on the tolerance of the human eye to the sequential color artifacts. The improved color management method is applicable to a variety of imaging systems including digital and analog imaging systems. Improved color management methods are also suitable for imaging applications other than display. These imaging systems can be front projectors and rear projection TVs. Such imaging systems can be used with an array of independently addressable pixels such as deflecting reflective micromirrors, mobile reflective films (eg IM〇D), liquid crystal cells (eg LCD), liquid crystal based units (LCOS), or radiation The system of the image engine of a tube (such as a plasma tube). These imaging systems may also include a passive or an active display. The improved color management method is also applicable to an imaging system using a sub-pixel space mosaic model, which is in the form of red, green and blue primary colors together with a colorless 2 pixel, using only achromatic sub-pixels which appear as an achromatic color on the screen; Or in the form of an ideal combination of the remaining color sub-pixels to enhance the glare reproduction and increase the brightness of the white. In addition, the improved color management can be used alone or in conjunction with not only a colorless sub-pixel or portion but also a sub-pixel or portion of its color. Referring to the drawings, Figure 1 is a structural diagram with multiple inputs. The color engine 110 has an integer that accepts - has - the color image component set {Ci} (i - to N) of the wheel image image component - „the same as the input image primary color number, for example, the wheel The input color set {Ci} may include red_blue' or other colors such as interlaced color difference elements (YCbCO, progressive scan color difference 7L pieces (YpbPr) and multimedia video color difference elements (γυν). " and color image 7L The input #{Ci} is converted into a _ color image component output set {Pk} (k 1 to M), wherein the total number of _ converted primary color image elements, Μ is greater than N. The color is even + jtm 4r , Baxin W engine produces one or more additional color components not included in the input set {C,}. Such a transformation can be expressed as: {(5)/;({(:,})' where rush, state, pu [ Eight (7), and the ^^ 乂 equation ^ ft is a conversion function, which can be a linear or non-linear function. One of the nonlinear forms of the function ft is Min(Xj), which takes the minimum value of the variable set & Another linear form of & can be: /(~)= Σ...(Equation 2) /=1 where rii is a set of coefficients; \ is a variant. In an example, after conversion The {pk} group has all the colors of the input color image element {C,}, and the extra color of the additional color image element is a linear combination of the additional color of the input color image element. For example, the input color set (Ci) can be Including (:1 = red image component, C2 = green image component, color set of blue image component. After conversion, the output color set {pi} can be package i27907.doc 1364713 Ρι = red image component, P2 = green image component, p3 = blue image component, and P4 = color set of white image component. In other examples, Pi can be any ideal color based on the input color set. For example, composing an output color set Colors can be selected in red, green, blue 'cyan blue, magenta, yellow and white. Note that 'although in different video, display applications and standards in white color" have different specifications' in this article" White " is not limited to a particular whiteness in a particular standard. Conversely, this "white, can be any particular "whiteness" in the current video and surveillance system. For example, when this application "White" When used as part of a rotating color disk, the white portion will still filter the incident light: that is, the outgoing (filtered) light emerging from the white portion has a spectrum different from the spectrum of the incident light. The saturation of the image element {Ci} or the energy it carries is redistributed between the output color image elements {pk}. Specifically, the saturation of the output additional color elements (or the energy they carry) is input from the input. The color image elements are allocated; the saturation of the input color elements is reduced. Thus, other output color image elements having the same color as the input color image elements will have lower saturation and energy than the input color image elements. As an aspect of this example, the conversion of at least one additional color image element includes a non-linear conversion function. For example, in an 8-bit image, the input image contains red, green, and blue image elements with a color magnitude of red = 255, green = 2 〇, and blue - 1G. By selecting [the minimum value for the input color image component; the nonlinear conversion function as the power law; the extra color is white, the round outset can include red, green, blue and extra color white, the values are red 127907 .doc 1364713 =254.9706, green=19.97〇6 'blue=9 97〇6, and white=_. In this example, white is a linear combination of substantially the same amount of red, green, and blue; the color magnitude is obtained by the -linear power law conversion function, which will be discussed later. In the above example shown in Fig. 1, the input and output ends of the color engine are parallel and the input color image elements can be transmitted in parallel to the color engine; the converted output color image elements can be output in parallel. In the alternative, the input and/or output of the color engine may be in series, which is not shown in the drawings. The improved color management method has many uses, among which - is applied to - imaging systems. An example of such an imaging system is shown schematically in FIG. Referring to Fig. 2, the imaging system 100 includes an image engine 1〇8, (4) learning to generate images from image data obtained from an ideal image. The image engine may include an array of independently addressable pixels, such as a deflecting reflective micromirror, a mobile reflective thin film (such as m〇D), a liquid crystal cell (such as an LCD), and a Shiyake liquid crystal cel. Or a tube (such as a plasma tube). The image engine may also include a sub-pixel spatial mosaic model, which is in the form of red, green and blue primary colors and a colorless sub-pixel: using only the colorless sub-pixels, which appear on the screen as achromatic; or the rest The ideal combination of color sub-pixels is used to enhance highlight reproduction and increase white brightness. The image f used by the imaging engine to generate the ideal image is modulated by the image processor 1-6 of the system controller 104, as shown in FIG. Specifically, the image processing H extracts (or receives) the input image signal image source from the image source i 02# may or may not be a component of the imaging system; then processes the input image 127907.doc 1364713 5 The appropriate format of the processed image, such as the bit-plane data format = can be used by the image engine. The alpha system controller is designed for the operation of the system features. In the image processor this is an example of a good color management method that can be implemented. Detailed statement: The image processor includes a color engine 11. As described above with reference to FIG. 3 - an example FIG. 3 schematically shows the junction of the image processor 1 〇 6 in FIG.

Referring to Fig. 3', image processor 1.6 includes a color engine " and a data formatter for converting image data from - format to a suitable capital format that can be used directly by the image engine. For example, the 'f formatter is suitable for converting pixel-by-pixel data (pixel data) into bit-wise data (bit-plane data). In one operation, the input image signal is converted by the color engine 11 into a converted round image signal {Pk}' which is further converted by the data formatter into a bit face data format. FIG. 4 is a schematic diagram showing another example of an image processor. Referring to FIG. 4, the image processor includes a color space converter 114, a color multiplexer 116, a color engine 110, and a data formatter 112. The color space converter (114) is designed to map the color space of the input image to the color space of the imaging system (1 〇 圆 in circle 2). Color multiplexer 116 outputs color image elements based on the output of the input image and color space converter (114). The processed image signal is passed to a color engine 110 for processing color images, which will be discussed below. The processed color image at the color engine 11 is then transferred to a data formatter 用于2 for converting the image data into a suitable format such as the bit face data that can be directly used by the image engine. 127907.doc 13: Other alternative configurations can be used for other functional modules, such as a grayscale correction (de gamma) module, such as between a color multiplexer and a data formatter. The above-mentioned color shirt engine (11 G) converts the input color image element into a new color image set. Using a specific example, the input color image component contains red and green 'blue' output color image elements including red, green, blue, extra color, white, which will be discussed below. Those skilled in the art should understand that the following discussion is for the purpose of demonstration and should not be construed as a limitation. Other variations that do not depart from the spirit of the invention are also possible. For example, the input color image elements can be any color combination, and the preferred colors can be selected in red 'green' blue, cyan, magenta, yellow, and white. The converted color image elements can also contain different color combinations. The resulting additional color shirt does not have to be a white color, but instead can be any other color or any number of different colors. Referring to Fig. 5, by converting the function ft, an intermediate value w can be calculated from the input Ci including red, green, and blue. For example, ft is the minimum operation of the input difference, then W is the minimum value of the R (red), G (green), (blue) signal (step 118). Then using a non-linear conversion function f, the minimum value w is converted to W (step 12 〇). Figure 7 shows three examples of nonlinear conversion functions that can be used to improve color management methods. Of course, other non-linear functions can also be used, such as the following equation: 丫 ·.. (Equation 3) where τ and ζ are variables. 127907.doc then subtracts the converted w from R, G, B to obtain new primary color values for R·, g, and b (step 122), as shown below. W -f(W)=f[fts(Rj 〇, B)]=f[Min(R, G, B)] G'=GW' As an example, the pixel values of the input image are r=255 , G=20, and B=1〇. w is set to 3〇, which is the minimum value of R, g, and b. Using the power law function shown in Figure 7, Wi can be calculated as 〇 294; R ′ G' and B, calculated as Ri = 254.9706, G, = 19.9706, Β' = 9·9706. Then, W, 'R,, G, and B are calculated to independently drive the four color bands in the display operation. In the above demonstration process, in the case where the conversion function is linear, the relative saturated color (R=255, G=20 'β = 1〇) does not remove and transmit the minimum value (b=1〇) component to White band. Conversely, such relatively saturated colors are not diluted by white elements', which enables imaging systems in which white data is used to display only white signals and near-white colors' which greatly reduces the sequential color of color separation artifacts. image. The above method can be implemented as a software having computer executable instructions; the software can be stored in a medium readable by a computer. Alternatively, the above method can be implemented in an electronic circuit device included in the color engine discussed below with reference to FIG. Figure 6 shows an example of an electronic circuit. This electronic circuit can be a field programmable gate array (FPGA) or a dedicated integrated circuit. Referring to Figure 6, device 126 includes a plurality of inputs for receiving primary color elements 127907.doc • 15- 丄 364713 { I ρ ν . Although the input end can also be a parallel connection of the serial wheel end, the smart combination is not shown in the figure 128 connected to the input conversion 杈., and an intermediate value = color image element is converted into a color image element Wei. The minimum operation, then the intermediate value is the input value of the small value ft. The output of the ft function module is connected to the input for converting the input signal W to the new value w, the input of the nonlinear calculator fi3〇

Hey. In this example, the new value w is output to other functional modules such as a grayscale correction (de-gamma) module for further processing. The calculated W is injected into the multiplier by the coefficient aR, which is multiplied and transmitted to the subtraction point respectively connected to the wheel ML signal to produce the output R. ’ G. And B. . Output w. For example, the output white element ' is obtained by multiplying the result of W by the coefficient ω and w, which is multiplied by the result. The mathematical formula can be expressed as: W,= f(W)=f[ft(Ri,Gi,Bi)]=f[Min(R,G,Β W0 = «xWUxW' = c〇xMin(Ri,Gi, (8)+(4) 如啦,g,(10) R〇'=Ri-aRxW'

Go^Gj-aoxW'

Bo^Bj-aBxW The coefficients aR, 仏 and ... are used to control the amount of subtraction from independent primary colors. These coefficients can also be attributed to the color correction factor, which is used to correct the "white" defined by the white portion of the color filter and the other color portions of the color filter, and the resulting white between the η"white" Difference. When the difference is substantially zero or lower than the threshold value, the coefficients aR, aG and ~ can all be 1 〇. Otherwise, the coefficients can be used to adjust the level of the primary color separately, in this case, aR, 127907 .doc 1364713 ccg and αΒ may or may not be consistent, but preferably include 〇 and i. The coefficient (1) and the artificially predetermined and preferably dynamically adjustable coefficients may have a value of 1. In another case, The dimension lookup table (L〇〇k_UpTabie; LuT) can be used with the improved color management method. - In this case, the primary color input (such as red, green, and blue primary colors) produces a primary color output and a white color. This configuration and the three-dimensional look-up table allow W' to be precisely controlled. For example, W can be specified as the minimum of the input primary colors such as R, G, and B; R, G., and B can be calculated as the result of the nonlinear function application. , and then, from R, the original The value is subtracted. This in turn causes the modification of the White-0nly-white imaging system in which white is composed of unequal R, G and Β can be counted as w, the component 'is manufactured-to-pure The white component (white_〇nly_white) system is brighter white. In other words, R'G, part of the B component can be fed into the white component to produce enhanced full power brightness. Part B can be dynamically adjusted based on human tolerance to sequential color image perception. The functions used to determine the amount of these parts are as follows: W'=tx W' input .255 ) 255 [1 + e_ (rx(,1 (Equation 4) where τ and ζ are variables, specified to control the number of white tl pieces produced by the input (r, the input can be any of R, g, and b. It can be implemented in the form of a software module that can be stored in a computer; the software can be stored in a medium that can be read by a computer. In addition, the above method can also be included in the front reference picture. I27907.doc 17 1364713 Implemented in electronic circuit devices in color engine 110 Figure 8 shows a schematic diagram of an electronic circuit for The electronic circuitry may be programmable gate array (FPGA) is a dedicated integrated circuit or a field.

Referring to Figure 8, device 132 includes a plurality of inputs for receiving primary color elements such as hearts, 〇| and . Although the inputs are connected in parallel, they may be in series but not shown. The ft conversion function module, which can be a minimized function, is connected to the inputs to obtain an input minimum. The output of the minimization function module is connected to the input of a non-linear filter f for converting the input signal into a new value W' after conversion. The calculated w is injected into the multiplier and multiplied by the coefficients aR, (6) and ...; the results are respectively transferred to subtraction nodes respectively connected to the input Ri, gab# to produce an output R. , G. And B. . The math formula is as follows:

Wt = f(W)=f[ft(Ri> Gi} Bi)]=f[Min(Rij Gj, B〇] R0'=Ri-(xRxW, Ο〇'=〇ϊ-α〇χ W' Bo ^Bj-aBxW' The coefficients aR, aG&aB are used to control the number of subtractions from the individual primary colors. These coefficients can also be attributed to the color correction factor, which is used to correct the white portion of the color filter. The difference between the white "and the other primaries of the color filter produces the difference between "white". The calculated R.' G. and B. are transmitted to multiple multiplications ^ and are respectively coefficients βκ, β. βΒ is multiplied. The product is input to an adder and the W-phase port multiplied by the coefficient co. The sum is output as a white signal w. It can be expressed by a mathematical expression as follows : 127907.doc •18· 1364713 W' = f(W)=f[ft(Ri5 Gi5 Bi)] = f[]VIin(Ri5 Gj, Bj)] R〇=Ri-aRX W' G〇=Gj- aG>< W' Β〇=Βί-αΒχ W' W0=^xW'+c〇XW+|^Rx(RraRxWI)+pGx(GiaGxW,) + hx (Bi-〇iBxW')] Figure 9 shows one In the full-power white reproduction image (r, g, B) compared with white, "Chinese white is the complex of different in the equation 2 (four) ~ Function generation. It can be seen in ®8, w, equal to 255; when the curve γ=〇2 shows full power white, R, 'G' and think 51. When the curve γ=1, the full power white W is $ R, G and Β are 255. Obviously, the white color produced when the curve γ=〇2” is better when the white color produced by the t匕 curve γ_1 is reduced, which is a linear conversion function. In an example, the power enthalpy is preferably less than 1, more preferably less than 〇 8. The improved color management method can be implemented in an imaging system. Specifically, it can be implemented in a sequential color imaging system. Reduced or eliminated sequential color enamel artifacts appear in most current sequential color imaging systems. As a scorpion, Figure 1 shows a sequential color imaging system using a rotating color disk. 'Kato 1 0' imaging system 134 Included is an illumination system providing illumination light 1 36, which adjusts the incident light according to the image data to reproduce the ideal image of the image 1 1 G. and emits the adjusted light to the screen 148 for viewing the projection lens ~ ', ,, Ming The system also includes a light source 13 8 which can be a Solitary light or other sources of lunar radiation such as lasers and light-emitting diodes, optical integrators 1 40, I27907.doc 1364713 color disk 142, and concentrating lens 144 ❶ The image engine includes any _ / I» A form of independently addressable pixel array. For example, the pixel array can be a deflection/reflection micromirror, a liquid crystal cell, a germanium-based liquid crystal cell, and other suitable devices. The image data used by the system controller is just an image processor from the image source 1G2. The input image is obtained. The system controller is used to control and synchronize the operation of the imaging system features. The color disk M2 includes a primary color group and a white portion, as shown in FIG. Referring to Fig. 11', the primary colors in this special case include red, green, and blue, and there are white portions. Each color portion can have any appropriate angle. For example, red, green, blue, and white can be angled at 120, 90, 90, and 60 degrees, respectively. It should be noted that the graph u σ u i 1, can not show a number of feasible examples: In other examples, the red 'green and blue primary colors may not be as shown in the figure. In addition, the color (four) may have other primary color combinations, preferably the primary colors may be in red, green, blue, cyan, and leap years. Choose between red and yellow. In addition to red, green and blue primary colors, extra color < $ his color. The color-aware group can be white. It is noted by those skilled in the art that the above-described embodiments are illustrative, and that the invention can be implemented in other ways. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a color engine; FIG. 2 is a schematic diagram of an imaging system; FIG. 3 is a schematic diagram of an image system of the imaging system of FIG. , wherein the image processing pirate includes a color engine of FIG. 1; FIG. 4 is a schematic diagram of another embodiment of the imaging system image of FIG. 2, wherein the image processor of 127907.doc • 20. 1364713 includes a picture 1 Figure 5 is a flow chart for performing the steps of calculating an additional color component other than the primary color component in the input image; Figure 6 is a schematic diagram of an apparatus for processing an image; A set of non-linear conversion functions for determining additional color components and a set of primary color components for generating images; FIG. 8 is a schematic diagram of another device for processing images; FIG. 9 is a white representation of a full power white In the image (R, g, B) and white, the white component is generated by using a power law conversion function of a different power; Figure 10 is a schematic diagram illustrating a sequential color imaging system; Figure 11 is a sequential imaging for use in Figure 10. A schematic view of the disk embodiment of the color system 'comprising the disk as a white color portion, and a red' green, and blue primaries of the group. [Main component symbol description]

100 Imaging System 102 Image Source 104 System Controller 106 Image Processor 108 Image Engine 110 Color Engine 112 Data Formatter 114 Color Space Converter 116 Color Multiplexer 118 Obtaining W from Ci using ft 127907.doc 1364713 120 Use A non-linear function f converts W to W1 122 calculates R1 on the basis of W', and G' and B1 124 calculate W from. 125 Outputs R', G1, B' and W. 126 Electronic Circuit Equipment 128 Conversion Module 130 Nonlinear Filter 132 Electronic Circuit Equipment

134 Imaging Systems 136 Lighting Systems 13 8 Light Sources 140 Optical Integrators 142 Color Plates 144 Condenser Lenses 146 Projection Lenses

148 Screen -22- 127907.doc

Claims (1)

Patent application scope - a method for processing an image, comprising: No. 096150407 Patent application Chinese patent application scope replacement (ϊοι年1月) Received a first with _第-色巴杉j a second color image component set of the second color set of the image of the color image component set; wherein the second color set includes an additional color not in the first color set; wherein the additional color - Derivation of the image element comprises - non-linear transformation; and wherein the derivation of one of the additional color image elements comprises more than one linear transformation applied to each of the elements of the first color image set. The method of claim 1, wherein the other color image elements in the second color group are derived from the (four) color image elements of the first color group and the additional color image elements of the second color group . 3. The method of item 2, wherein the second set of color elements in the image: - the energy distribution is different from the energy distribution of the first set of color elements in the image. 4. According to the claim 2, the second color group contains all the colors of the first sound group. & 5. The method of claim 4, wherein the first color group comprises red 'green and blue'; and the poor music color set includes red, green, blue, and white. 6. According to claim 3, the additional color component is generated by a program, the program comprising: obtaining an intermediate value from the first Xincai-VaL I component set; 127907-10101H.doc Applying the non-linear function to the intermediate value of the acquisition to obtain an intermediate value; and, setting the additional color element is equivalent to the converted intermediate value. The method of claim 7, wherein the intermediate value is the minimum value of the first color element, according to the method of claim 7, further comprising: subtracting the converted color from each color component of the first color component set An intermediate value to obtain each color element of the second set of color elements. According to the method of claim 6, the non-linear function of the towel is a power law function whose power is less than 〇 8. An imaging system for reproducing an input image, comprising: an image processor having an input end for receiving an input image to be reproduced, wherein the image processor further comprises: an input terminal for receiving the a color component set of the input image; a first conversion module 'for obtaining an intermediate value from the color component set; β a nonlinear filter connected to the first conversion module and using a nonlinear function to convert One of the first conversion modules outputs a signal; a subtraction node and a multiplexer group are connected to the input terminals of the first conversion module and the output end of the nonlinear filter to obtain a converted color a set of image elements; and one or more linear filters 'connected to one of the outputs of the non-linear filter', an output of the first conversion module, and for converting an output of the first conversion module One of the subtraction node groups; and 127907·101 〇1 ί | .doc -2- an output for outputting the converted color image component set; One originated from the turn The subsequent color image element 1 further includes an output terminal of the nonlinear filter and an output end of the first conversion module according to the line of the request object; and the output unit has an output of the adder The output of the converted white image component. 12. An apparatus for processing an image, comprising: an input for receiving a set of color components of an input image; a first conversion module for obtaining an intermediate value from the set of color components; a nonlinear filter connected to the first conversion module to use a nonlinear function to convert the intermediate value; a subtraction node group connected to the input terminals of the first conversion module and the output of the nonlinear filter End, subtracting a value proportional to the nonlinear conversion value from each individual input color image 7L piece t; or a plurality of linear filters connected to one of the output ends of the nonlinear filter for each individual input A color image element is subtracted from a value proportional to the non-linear conversion value; and an output signal set is used to output a set of converted color image elements including the subtracted color image elements. 13. The device of claim 12, wherein the converted set of color image elements comprises an additional color image element that is different from any of the input color image elements. 127907-10101. The device of claim 13, wherein the additional color image element is a one of an adder that rotates a signal, the adder having a first connection to the first output of the first conversion module - an input; and - a second input connected to the output of the non-linear filter. The first input end of the adder is connected to the output end of the first conversion module by a multiplexer, and the multiplexer is adapted to transmit the first conversion mode by using - coefficient multiplexing The output signal of the group. Root: the device of claim 15 wherein the second input of the adder is connected to the output of the non-linear filter through a further multi-protector, the multiplex write = transmitting the nonlinear chopping with another coefficient multiplication The output signal of the device. The apparatus of claim 15, wherein each of the subtraction nodes has an output for rotating the converted color image elements. 127907-1010111.doc